Summary
2007 International Symposium on Nonlinear Theory and its Applications
2007
Session Number:18AM1-A
Session:
Number:18AM1-A-4
Sparse and Passive Reduced-Order Interconnect Modeling by Eigenspace Method
Yuichi Tanji,
pp.120-123
Publication Date:2007/9/16
Online ISSN:2188-5079
DOI:10.34385/proc.41.18AM1-A-4
Summary:
The passive and sparse reduced-order modeling of RLC networks is presented, where eigenvalues and eigenvectors of the original networks are used, thus, the macromodels obtained from the proposed method are more accurate than the Krylov subspace methods and TBR procedured to a class of problems. Furthermore, the proposed method is applied to post-processing after a reduced order model is obtained from the Krylov subspace methods of TBR procedures without breaking the passivity condition and losing accuracy of the marocmodels. Therefore, the proposed eigenspace method is not only an alternative of other reduced-order macromodeling methods, but also is embedded in these methods enhancing their performances.